JP2020038488A - Case occurrence suppression effect prediction system - Google Patents

Abstract

To effectively determine a warning destination by predicting a suppression effect of a case occurrence which is caused by a warning of a police officer.SOLUTION: A case occurrence suppression effect prediction system includes: a database for storing data to be used to predict a case occurrence; a learning server for learning the data stored in the database to construct a prediction model 1400 of the case occurrence; a predicting server for predicting a suppression effect amount of the case occurrence with reference to the prediction model 1400; and a user terminal for visualizing and displaying the prediction effect amount predicted by the predicting server on a map with reference to input data inputted by a user. The learning server learns geographic information and location information of a police officer with the existence/nonexistence of the case occurrence as teacher data to construct the prediction model 1400 of the case occurrence, the predicting server calculates a suppression effect amount of the case occurrence by using the prediction model 1400 constructed by the learning server and the input data inputted from the user terminal, and the user terminal visualizes and displays the suppression effect amount of the case occurrence on the map.SELECTED DRAWING: Figure 14

Description

  The present invention relates to a case occurrence suppression effect prediction system.

  A technique has been proposed in which a template matching method is applied to person position statistical information created based on position information and attribute information of each person to predict the place and time of occurrence of a crime (Patent Documents) 1).

  In addition, a crime prediction system that determines a crime prediction by adjusting a past crime rate based on a correlation between weather conditions and crime data has been proposed (see Patent Document 2).

Japanese Patent Application Laid-Open No. 2006-166938 (Japanese Patent No. 5992563) JP-T-2018-505474

  However, if it is supposed to prevent the occurrence of a case beforehand such as a police officer's guarding activity, it is necessary to consider the deterrent effect on the occurrence of a case caused by a police officer guarding. If the location of the police officer is determined without being able to estimate the effect of the police officer, there is a possibility that a place with a high deterrent effect on the occurrence of a case may be overlooked.

  Patent Documents 1 and 2 do not consider a deterrent effect on occurrence of a case caused by a police officer guarding.

  SUMMARY OF THE INVENTION An object of the present invention is to effectively determine a point to be watched by predicting a deterrent effect of occurrence of a case caused by a police officer watching.

  An incident occurrence suppression effect prediction system according to one embodiment of the present invention includes a data server having a database that stores data used for predicting the occurrence of a case, and learning the data stored in the database to predict the occurrence of the case. A learning server for constructing a model, a prediction server for predicting the amount of suppression of the occurrence of the case by referring to the prediction model, and a prediction server for predicting the effect amount of the case occurrence by referring to input data input by a user. A user terminal that visualizes and displays the predicted effect amount on a map, and the learning server learns the geographic information and the position information of police officers as teacher data on whether or not the case has occurred. Constructing the prediction model of the case occurrence, the prediction server calculates the prediction model constructed by the learning server and the input data input from the user terminal. There seeking the deterrent effect of the cases occurs, the user terminal is visualized and displaying before article proposed the deterrent amount the on the map generation.

  According to an aspect of the present invention, a destination to be watched by a police officer can be effectively determined by predicting a deterrent effect of occurrence of a case caused by watching by a police officer.

1 is a schematic diagram of an overall configuration of a case occurrence suppression effect prediction system according to a first embodiment. (A) is a diagram showing a functional configuration of a learning server, and (b) is a diagram showing a functional configuration of a prediction server. It is a time-sequence sequence diagram showing the exchange between each server of a case occurrence suppression effect prediction system. It is a figure which shows typically an example of the input table for learning used for construction of a case occurrence prediction model. It is a flowchart which shows the learning process for producing the prediction model which outputs a case occurrence probability. It is a figure which shows typically an example of the input table for prediction used for calculating the case occurrence probability according to the presence or absence of Police, etc., (a) is a schematic diagram in the case of carrying out Police, and (b) ) Is a schematic diagram in a case where the guard is not implemented. It is a flowchart which shows the prediction process for estimating the incident occurrence suppression effect amount by a policeman. FIG. 3 is a diagram illustrating a graphical user interface of the incident occurrence suppression effect prediction system. FIG. 14 is a diagram schematically illustrating an example of a case priority weight table defining a priority case type according to the second embodiment. FIG. 14 is a functional configuration diagram of a prediction server 120 to which a case priority weight storage unit 1000 according to a second embodiment is added. 9 is a flowchart for calculating the alarm priority for determining the alarm destination according to the second embodiment. FIG. 13 is a diagram schematically illustrating an example of a learning input table format used for constructing a case occurrence prediction model including a watch type variable according to the third embodiment. 14 is a flowchart illustrating a prediction process for predicting a case occurrence suppression effect amount for each type of alarm according to the third embodiment. It is the whole flow schematic diagram which represented the whole flow of learning and prediction typically. FIG. 14 is a diagram schematically illustrating an example of a table defining a pattern of the embodiment of the watch of Example 3;

  Hereinafter, embodiments will be described with reference to the drawings.

With reference to FIG. 1, an overall configuration of a case occurrence suppression effect prediction system according to a first embodiment will be described.
The incident occurrence suppression effect prediction system according to the first embodiment basically includes a data server 100 having various databases used for prediction, a learning server 110 for learning data to construct a prediction model, and a case occurrence suppression effect prediction. And a computer 130 (user terminal) having an application for visualizing a prediction result on a map.

  The main databases used in the case occurrence suppression effect prediction system include a case database 101 storing information such as date and time of occurrence, a place, and a case type, and a geographic information database 102 storing information such as demographics and facilities. And a GPS database 103 storing position information of police officers and a map database 104 storing map information for visualization. However, besides these, for example, there may be a demographic database or the like based on location information of a mobile phone.

  The case database 101 stores at least information such as a case ID, a date, a time, a place, and a case type at which the case occurred. The case type includes, for example, a traffic accident, a murder, a burglary, a bicycle theft, and the like.

  The geographic information database 102 stores demographic data including population ratios and household ratios by age, the location and use status of facilities (for example, police boxes, police stations, stations, schools, etc.), and land use classifications (for example, residential areas). , Fields, roads, forests, etc.). Further, if it can correspond to geographical position information, it may include information such as weather, temperature, precipitation, occurrence of an event such as sports, and the like.

  The GPS database 103 stores position information obtained by a position measurement function of a terminal carried by a police officer and a device mounted on a vehicle. This position information includes at least time stamp, latitude, and longitude information. Further, in addition to these three elements, information such as a police officer's business form and a vehicle type may be included.

  The map database 104 is map data for a geographic information system (GIS) used for displaying a predicted location of a case on an application. The map data does not necessarily need to be possessed in the form of the map database 104 as long as it can be separately acquired online.

  The learning server 110 and the prediction server 120 receive necessary data from the data server 100, process the data into an input table format, and then perform a calculation process by machine learning. The learning server 110 learns the geographic information and the police position information data using the incident as teacher data to construct an incident occurrence prediction model. The prediction server 120 finally outputs a prediction result of the case occurrence suppression effect using the case occurrence prediction model constructed by the learning server 110 and user input data transmitted from the computer 130 used by the user. Thus, the user can confirm from the computer 130 a place where the case occurrence suppression effect is predicted to be high.

The functional configuration (a) of the learning server 110 and the functional configuration (b) of the prediction server 120 will be described with reference to FIG.
As shown in FIG. 2A, the learning server 110 processes the input data into an input table format as shown in FIG. 4, and an input unit 200 for inputting data obtained from the data server 100. A data processing unit 201, an input table storage unit 205 for storing an input table, a learning control unit 202 for controlling the timing of performing learning, a prediction model construction unit 203 for performing learning, and outputting the constructed prediction model And a prediction model storage unit 206 that stores a prediction model.

  In order to predict the case occurrence suppression effect for each time and place, the data processing unit 201 aggregates various data for each prediction unit. As for the location, a geographic range to be predicted is meshed, and an ID is assigned to each mesh. The mesh is divided into, for example, a square mesh having a side of 100 m, and the mesh size may be larger or smaller. For each set mesh, a place where the case occurrence suppression effect is high / low is predicted. In the same manner, the time is predicted in a certain time zone size. For example, a prediction every two hours is taken as an example, but the width of two hours is merely an example, and may be longer or shorter.

  Next, as illustrated in FIG. 2B, the prediction server 120 includes an input unit 210 that receives data acquired from the data server 100 and a prediction model acquired from the learning server 110, and a prediction request from the user. A request receiving unit 215 that receives input and input parameters, a data processing unit 211 that processes data and input parameters into an input table format as shown in FIG. 6, a prediction control unit 212 that controls prediction execution for each prediction condition, A prediction processing unit 213 that performs a process of suppressing effect prediction using the prediction model and the input table received from the learning server 110, an output unit 214 that outputs a prediction result, and a prediction result storage unit 216 that stores the prediction result. , Is provided.

  It is not always necessary to execute the prediction process each time a prediction request is received from the user, and the prediction result stored in the prediction result storage unit 216 may be output for a condition that has already been predicted.

Next, the flow of exchange between the servers will be described with reference to the time-series sequence of FIG.
As shown in FIG. 3, the learning server 110 receives data from the data server 100 as input, executes a learning process, and outputs a prediction model. The prediction server 120 receives as input a prediction model from the learning server 110, data from the data server 100, and input parameters from a computer 130 operated by a user. The predicted value of the deterrent effect is calculated using these inputs, and the predicted result is output to the computer 130 side.

  Since learning is a process with a high time cost, basically, after a prediction model is constructed, a prediction process is executed using the same prediction model for a prediction request from a user. However, the learning control unit 202 may execute re-learning when a decrease in prediction accuracy is confirmed or at a regular timing.

Based on the above premise, a method of predicting the case occurrence suppression effect will be described with reference to the overall flow of FIG.
As shown in FIG. 14, in the learning, first, various data are processed to create a learning input table, and the relationship between the past geographic information and the execution status of the guard and the occurrence of a case are learned by machine learning. By this learning, a prediction model 1400 is constructed and output.

In the prediction, at the future date and time, the case occurrence probabilities for each mesh in the case where the guard is performed and the case where the guard is not performed are respectively predicted. These are output by creating two types of prediction input tables in which only the values of the variables are changed and inputting them to the prediction model 1400. Finally, the difference in the probability of occurrence of the case due to the difference in the presence or absence of the police is regarded as the deterrence by the police guard, and this is output as the deterrent effect amount.
The individual details of the overall flow will be described below.

FIG. 4 is a diagram showing an example of a format of a learning input table used at the time of learning.
As shown in FIG. 4, the learning input table holds a mesh ID, a date, a time zone, each geographic information variable group, an alarm variable, and a case variable as columns. Each geographic information variable is information associated with a mesh ID, a mesh ID and a date, or a mesh ID and a time zone, and is acquired from the geographic information database 102. The case variable is teacher data having a value of 1 if a case has occurred and a value of 0 if not occurring in each mesh ID, date, and time zone. The presence / absence of a case is calculated based on information on the date and time of occurrence of each case in the case database 101 and information on the location.

  The alarm variable is a variable having a value of 1 if the alarm was implemented before the occurrence of the case and a value of 0 if the alarm was not implemented in each mesh ID, date, and time zone. The presence or absence of the police is collected from the position information of the police officer or the vehicle in the GPS database 103 and the time stamp information. Note that the mesh ID and the date are not used as feature amounts in machine learning.

  A process of constructing a prediction model 1400 by machine learning in the learning server 110 will be described with reference to the learning flowchart in FIG. In this learning, a model for predicting the future occurrence of a case is constructed from geographical information and enforcement information using the presence or absence of a case as teacher data.

  First, the input unit 200 acquires the case, geography, and GPS information from each database (S500). When the data is obtained, the data processing unit 201 processes the data into the format of the learning input table shown in FIG. 4 (S501).

  Next, when the creation of the learning input table is completed, the prediction model construction unit 203 performs the prediction model construction by machine learning by using the case variables as the teacher data, the time period, the geographic information, and the alarm variable as the feature vectors. Perform (S502). A logistic regression model or the like can be used as an example of the machine learning algorithm here. However, the algorithm is not limited to logistic regression as long as the algorithm can express the output as a probability value. By this processing, when a feature value of a certain mesh in a certain time zone in the future is given, a case occurrence prediction model that outputs a probability value that is classified into a case occurrence class is generated.

  Finally, the output unit 204 stores the case occurrence prediction model and outputs the model to the prediction server 120 (S503).

Subsequently, a method of predicting the case occurrence suppression effect in the prediction server 120 will be described. FIG. 6 is a diagram showing an example of the format of a prediction input table used at the time of prediction.
Like the learning input table (see FIG. 4), the prediction input table holds a mesh ID, a date, a time zone, each geographic information variable group, and an alarm variable as columns. The case variable has an unknown value at the time of prediction and does not exist in the prediction input table. Two types of input tables for prediction having the above-mentioned format are prepared: a case where the guard is performed (a) and a case where the guard is not performed (b). The mesh ID, date, and time zone are the value of the place to be predicted and the date and time, respectively, and are set to the same value in the two input tables for prediction. Since each geographic information variable is also a value corresponding to a place and a date and time, the two prediction input tables have the same value.

  What differs between the two input tables for prediction is the value of the alarm variable. In the prediction input table (a), which is a case where the alarm is performed for the place and the date and time to be predicted, all 1 are stored in the alarm variables. On the other hand, all 0s are stored in the alarm variables in the input table for prediction (b) in which alarm is not performed. Using the two types of prediction input tables created here, the deterrent effect by the police is predicted.

With reference to the flowchart at the time of the prediction in FIG. 7, a process of predicting the effect of suppressing the occurrence of a case by the machine learning executed by the server for prediction 120 will be described.
First, the input unit 210 acquires a prediction model of a case occurrence from the learning server 110 (S700). Further, the request receiving unit 215 acquires input parameters from the computer 130 terminal operated by the user (S700). The input parameters include a date and a time zone in which the user wants to perform a forecast, forecast weather information in the time zone, a case type to be predicted, and the like. However, the input by the user may be omitted by automatically acquiring the weather and the like online. Then, the input unit 210 acquires the case, geography, and GPS information from each database again (S700).

  Next, the data processing unit 211 creates the two types of input tables for prediction shown in FIG. 6 (S701). After the prediction input table is created, the prediction processing unit 213 inputs the two types of prediction input tables to the prediction model, respectively, and the probability of occurrence of a case when the guard is performed in each mesh and the case that the guard is not performed in each mesh. The probabilities are calculated respectively (S702). Then, the difference in the probability of occurrence of a case depending on whether or not the guard is implemented is calculated, and this is defined as the amount of deterrent effect by the guard (S702). Note that the suppression effect amount may be defined as not only the difference but also a decrease rate of the probability of occurrence of a case from when the alarm is not performed to when the alarm is performed.

  Finally, the output unit 214 stores the suppression effect amount and outputs it to the computer 130 terminal used by the user (S703).

FIG. 8 is an example of a user interface on an application on the computer 130 operated by the user.
As shown in FIG. 8, a map 800 acquired from the map database 104 is displayed on the screen. The user selects a date selection button 801 for selecting a date on which the case occurrence suppression effect is to be predicted, a time period selection button 802 for selecting a time period for which prediction is to be performed, a weather selection button 803 for selecting forecast weather information at the date and time to be predicted, A case selection button 804 for selecting a case type to be predicted is pressed and set, and a prediction display button 805 is pressed. Then, the prediction result of the suppression effect amount for each mesh is output from the prediction server 120, and the prediction result 806 for each mesh is expressed in shades of color and visualized on the map 800.

  Further, the map can be enlarged and reduced by the scale setting bar 807, and the mesh size is also enlarged and reduced in accordance with the enlargement and reduction of the map.

  Since the prediction result 806 for each mesh is determined based on the predicted value of the suppression effect amount, the shading of the color may be displayed in a plurality of stages. Further, the predicted value may be directly displayed on each mesh. Also, the predicted value may be displayed in a balloon only when the mouse cursor is placed on the mesh.

  The mesh size may be dynamically changed in accordance with the enlargement and reduction of the map. For example, when the map 800 is reduced, the prediction result 806 for each mesh may be displayed by aggregating the mesh size to 500 m square. . In this case, the predicted value may be a final predicted value, for example, by averaging the predicted values of the 100 m square mesh included in the 500 m square mesh.

  Regarding the case selection button 804, a plurality of case types may be selectable. In such a case, for example, the predicted values of the suppression effect amount for each case type may be summed up for each mesh and expressed in shades of color. In that case, for example, when the mouse cursor is placed on the mesh, the predicted value for each case type may be displayed in a balloon. In addition, not only the suppression effect amount but also the case occurrence probability calculated in the derivation process may be visualized and confirmed on another screen.

  According to the first embodiment, each police officer can confirm a place where the case occurrence suppression effect is high by the police through the computer 130, and can determine the destination of the police based on the information. Thus, it is possible to take a precautionary measure at a place where the deterrent effect is expected to be high based on the relationship between the past police results and the occurrence of the incident, and to take a precautionary measure, thereby preventing the occurrence of the incident.

  A case occurrence suppression effect prediction system according to the second embodiment will be described. The second embodiment is a method in which when the suppression effect amount for each mesh in the first embodiment is calculated for each case type, the priority of the guard for determining the guard is considered.

  FIG. 9 is a case priority weight table in which priority weights are set for each case type. The user sets the priority for each case type as a weight. For example, in the example of FIG. 9, a high weight is set for “murder”, which means that the priority of “murder” is higher than that of other case types.

  Second Embodiment A prediction server according to a second embodiment will be described with reference to FIG. The prediction server according to the second embodiment is obtained by adding a case priority weight storage unit 1000 to the prediction server 120 according to the first embodiment (see FIG. 2B).

  A case priority weight table is stored in the case priority weight storage unit 1000, and the prediction processing unit 213 calls the case priority weight table after the amount of suppression effect for each case type is calculated.

  With reference to the flowchart at the time of prediction in FIG. FIG. 11 is a flowchart for calculating the alarm priority for each mesh by adding a new process to the first embodiment.

  Only the parts that are different from the first embodiment will be described. First, the request receiving unit 215 obtains the case priority weight from the user together with the input parameters, and stores the case priority weight in the case priority weight storage unit 1000 in the form of a case priority weight table (S1100). Then, after the prediction processing unit 213 calculates the suppression effect amount for each case type, the prediction processing unit 213 acquires the case priority weight table from the case priority weight storage unit 1000.

Then, the suppression effect amount for each case type is multiplied by the priority weight of the corresponding case type in the case priority table. The result of the multiplication of the suppression effect amount and the priority weight is defined as an alarm priority for each mesh (S1101).
Finally, the output unit 214 saves and outputs the alarm priority (S1102).

  As described above, according to the second embodiment, by assigning the case priority weights, for example, when there are two meshes having the same suppression effect amount, the one with the more important case type is presented with priority. Can be. Also, for example, when the suppression effect amounts for a plurality of case types are calculated for each mesh, the results obtained by multiplying each case by the case priority weight are summed up. You can decide ahead.

  A case occurrence suppression effect prediction system according to the third embodiment will be described. Third Embodiment The third embodiment extends the suppression effect amount calculation based on the difference in the presence or absence of the execution of the guards in the first embodiment to the suppression effect amount calculation based on the difference in the embodiment of the guards. Here, the police officer's embodiment is information defined on the basis of police officer's implementation means, such as the number of police officers at the time of police officers and the use of vehicles.

  FIG. 15 is a diagram showing an example of a pattern of the embodiment of Police. Based on the information contained in the GPS database 103, combinations of the number of guards and the use or non-use of vehicles are defined as an embodiment pattern, and code values are assigned. If another embodiment pattern can be defined from the GPS database 103, a pattern other than the number of guards and whether or not the vehicle is used may be defined. It should be noted that the alarm type includes the case where no alarm is performed, and here, the code value is set to 0.

  FIG. 12 is a diagram in which the alarm variable in the learning input table according to the first embodiment is changed to an alarm type variable. In the policeman variable, the presence or absence of the policeman is expressed in binary, but the policeman type variable is represented by a code value defined in the policeman's embodiment pattern. In each past mesh ID, date, and time zone, the embodiment pattern of the policeman actually applied at that time and place is set as the value of the policeman type variable. The pattern of the embodiment is included in the feature quantity as a type variable, and the relationship with the occurrence of a case is learned.

  Next, with reference to the prediction flowchart of FIG. 13, the calculation of the suppression effect amount due to the difference in the type of the alarm will be described. Here, FIG. 13 is a prediction flowchart of the deterrent effect amount for each type of alarm. Only the parts that are different from the first embodiment will be described.

  The request receiving unit 215 obtains a security resource based on the above-described embodiment pattern from the user (S1300). The watchman resources refer to the number of people, the number of vehicles, and the like that the user can assign to the watchman on the date, time zone, and geographic range to be predicted. For example, when the embodiment pattern is defined as a combination of the number of people and whether or not the vehicle is used, the user inputs the number of people and the number of vehicles as an alarm resource.

  Subsequently, the data processing unit 211 determines an executable embodiment pattern from the alarm resource, creates an alarm type variable having each embodiment pattern as a value, and inputs a prediction input table for executing a specific alarm type. Is created (S1301).

  As an example, consider the case where the police resources are two police officers and one vehicle. Executable alarm types are five types as shown in FIG. As in the first embodiment, five prediction input tables are created in which only the value of the type variable is changed. The prediction processing unit 213 inputs the five prediction input tables to the prediction model and calculates the case occurrence probabilities. In other words, it means that the likelihood of occurrence of a case when each type of police officer is implemented is predicted for each type of police officer.

As in the case of the first embodiment, the probability of occurrence of the deterrent effect in the case where the type variable of the alarm is “no alarm” and a specific value other than that (for example, “two police officers, vehicle is present”). Calculate the difference of
As a result, the amount of deterrence effect in the case where a specific type of police officer (in the above example, “two police officers, vehicle present”) is implemented. This is calculated for each type of alarm (S1302).

  In the third embodiment, if the amount of deterrent effect can be predicted for each type of police, for example, in the same mesh, if it is determined from the prediction result that the deterrent effect does not change even if the number of police officers corresponding to the police is increased, the resources of the police are saved. It is possible to carry out efficient alarm.

  As described above, in the above embodiment, by learning the relationship between the past crime occurrence and the enforcement of police officers, it is possible to predict the case occurrence suppression effect by the police officers at the place where police officers are to be implemented from now on. The location can be presented. That is, by predicting the case occurrence suppression effect of each place by the police, an effective guard can be determined based on the magnitude of the case occurrence suppression effect.

Reference Signs List 100 data server 101 case database 102 geographic information database 103 GPS database 104 map database 110 learning server 120 prediction server 130 computer (user terminal)

Claims (11)

A data server having a database for storing data used for predicting the occurrence of a case,
A learning server that learns the data stored in the database and builds a prediction model of the case occurrence;
With reference to the prediction model, a prediction server for predicting the suppression effect amount of the case occurrence,
A user terminal that refers to input data input by a user, visualizes and displays the predicted effect amount predicted by the prediction server on a map,
The learning server,
Learning the geographic information and the position information of police officers as the presence or absence of the case occurrence as teacher data, constructing the prediction model of the case occurrence,
The prediction server,
Using the prediction model constructed by the learning server and the input data input from the user terminal to determine the suppression effect amount of the case occurrence,
The user terminal,
A case occurrence suppression effect prediction system, wherein the suppression amount of the case occurrence is visualized and displayed on the map. The database is
A case database that stores information on the date and time of occurrence, place and type of case;
A geographic information database storing the geographic information;
A GPS database for storing the position information of the police officers,
A map database for storing map information,
The case occurrence suppression effect prediction system according to claim 1, comprising: The learning server,
It has a learning input table that holds mesh ID, date, time zone, geographic information variable, alarm variable and case variable,
The mesh is configured by meshing a prediction target geographic area,
The geographic information variable is information that associates the mesh ID with the date or the time zone,
The case variable is teacher data indicating whether or not the case has occurred in the mesh ID, the date, and the time zone, and the occurrence of the case is determined by the date and time of occurrence of the case in the case database. And location information,
The guard variable is a variable indicating whether or not the guard is performed before the occurrence of the incident in the mesh ID, the date, and the time zone, and the presence or absence of the guard is determined by the position information of the police officer in the GPS database. The case occurrence suppression effect prediction system according to claim 2, wherein the calculation is performed from the following. The prediction server,
It has an input table for prediction that holds the mesh ID, date, time zone, geographic information variables and alarm variables,
The mesh is configured by meshing a prediction target geographic area,
The geographic information variable is information that associates the mesh ID with the date or the time zone,
The guard variable is a variable indicating whether or not the guard is performed before the occurrence of the incident in the mesh ID, the date, and the time zone, and the presence or absence of the guard is determined by the position information of the police officer in the GPS database. The case occurrence suppression effect prediction system according to claim 2, wherein the calculation is performed from the following. The prediction server,
The prediction input table is created in two types: a case where a guard is implemented and a case where a guard is not implemented;
By inputting the two types of prediction input tables into the prediction model, a difference in the probability of occurrence of a case due to a difference in the presence or absence of a guard is obtained, and the effect of suppressing the occurrence of the case by performing the guard is performed. The case occurrence suppression effect prediction system according to claim 4, wherein the amount is obtained. The prediction server,
The suppression effect amount is obtained for each of the meshes,
The user terminal,
The incident occurrence suppression effect prediction system according to claim 5, wherein the suppression effect amount is displayed on the map for each of the meshes. The user terminal,
The map information obtained from the map database;
A date selection button for selecting a date for predicting the suppression effect amount of the case occurrence,
A time zone selection button for selecting a time zone for estimating the suppression effect amount of the case occurrence,
A weather selection button for selecting forecast weather information at a date and time for predicting the suppression effect amount of the case occurrence,
A case selection button for selecting a case type for estimating the suppression effect amount of the case occurrence,
Display prediction display button and
The prediction server,
In response to the press of the prediction display button, determine the suppression effect amount in the mesh,
The user terminal,
The incident occurrence suppression effect prediction system according to claim 6, wherein the suppression effect amount for each of the meshes is displayed on the map. The user terminal,
The case occurrence suppression effect prediction system according to claim 7, wherein the suppression effect amount for each of the meshes is expressed and displayed on the map in shades of color. The prediction server,
It further has a case priority weight table in which case priority weights are set for each case type,
For the deterrent effect amount for each of the case types, the result obtained by multiplying the case priority weight of the case type corresponding to the case priority table in the case priority table is obtained as an alarm, and
The user terminal,
The case occurrence suppression effect prediction system according to claim 1, wherein the priority is displayed. The prediction server,
The amount of the deterrent effect in the case of performing a specific type of police officer is obtained for each type of police officer,
The user terminal,
The case occurrence suppression effect prediction system according to claim 2, wherein the amount of suppression effect is displayed for each type of the alarm. The prediction server,
11. The combination of the number of police officers or the presence or absence of use of a police vehicle based on position information of police officers included in the GPS database as the police officer type is defined as a police officer embodiment pattern, according to claim 10. Incident deterrent effect prediction system.

Images